Authors:
André Ippolito
and
Jorge Rady de Almeida Júnior
Affiliation:
Polytechnic School of University of São Paulo, Brazil
Keyword(s):
Ontology Matching, Aspect, Consensus Clustering, Bayesian Cluster Ensembles, Community Detection.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Biomedical Engineering
;
Cloud Computing
;
Collaboration and e-Services
;
Complex Systems Modeling and Simulation
;
Coupling and Integrating Heterogeneous Data Sources
;
Data Engineering
;
Data Mining
;
Databases and Information Systems Integration
;
e-Business
;
Enterprise Information Systems
;
Health Information Systems
;
Information Systems Analysis and Specification
;
Integration/Interoperability
;
Interoperability
;
Knowledge Engineering and Ontology Development
;
Knowledge Management
;
Knowledge Management and Information Sharing
;
Knowledge-Based Systems
;
Ontologies and the Semantic Web
;
Ontology Engineering
;
Semantic Web Technologies
;
Sensor Networks
;
Services Science
;
Signal Processing
;
Simulation and Modeling
;
Society, e-Business and e-Government
;
Soft Computing
;
Software Agents and Internet Computing
;
Software and Architectures
;
Symbolic Systems
;
Web Information Systems and Technologies
Abstract:
With the increase in the number of existing ontologies, ontology integration becomes a challenging task. A fundamental step in ontology integration is ontology matching, which is the process of finding correspondences between elements of different ontologies. For large-scale ontology matching, some authors developed a divide-and-conquer strategy, which partitions ontologies, clusters similar partitions and restricts the matching process to ontology elements of similar partitions. Works related to this strategy considered only a single ontology aspect for clustering. In this paper, we proposed a solution for ontology matching based on Bayesian Cluster Ensembles (BCE) of multiple aspects of ontology partitions. We partition ontologies applying Community Detection techniques. We believe that BCE of multiple aspects of ontology partitions can provide an ontology clustering that is more precise than the clustering of a single aspect. This can result in a more precise matching.